Raghavendra

You all wanna see the future? We're bringing it to you. Here's SupraOS' Agent Orchestration Demonstration, real Supra Oracle data, real web search, real agent coordination. The next video we'll show you AI Agent DeFi Trading ;) Those of you that adopt tools like this are going to be MILES AHEAD of folks that don't. We've selected the first cohort of SupraOS Alpha users, you'll be notified soon. Your SUPERPOWERS await... p.s. there will be at least 5 ways $SUPRA will be used to power SupraOS.

Part 2 in the Crypto+Law blog series: How India has evolved / adopted crypto? Indian populace marching ahead of regulators! @raghavendra.kaundinya/crypto-the-transformative-power-of-decentralisation-channelled-through-governments-cae5312f4705" target="_blank" rel="nofollow noopener">medium.com/@raghavendra.k…

Crypto and blockchain are built on decentralisation. Governments and regulations are products of centralisation. The two should be opposites. Or are they? A small attempt at studying how the United States has answered that question: @raghavendra.kaundinya/crypto-the-transformative-power-of-decentralisation-channelled-through-governments-e6f885cd0313" target="_blank" rel="nofollow noopener">medium.com/@raghavendra.k…

🚨 Someone just turned your WiFi router into a full-body surveillance system. No cameras. No wearables. No video. Just radio waves. It's called RuView. It uses the WiFi signals already in your room to detect human poses, track breathing, measure heart rate, and see through walls. Not a concept. Not a research paper. Working code you can run right now. Here's what this thing actually does: → Tracks full 17-point body pose using only WiFi signals → Detects breathing rate (6-30 BPM) without touching anyone → Measures heart rate (40-120 BPM) from across the room → Sees through walls, furniture, and debris up to 5 meters deep → Tracks multiple people simultaneously with zero identity swaps → Self-learns from raw WiFi data. No labeled datasets needed Here's how it works: WiFi signals pass through your room and hit the human body. The body scatters those signals differently based on position, breathing, even heartbeat. RuView reads that scattering pattern and reconstructs everything. A mesh of 4 ESP32 nodes ($48 total) gives you 360-degree coverage with 12 measurement links, 20 Hz updates, and sub-30mm precision. Here's the wildest part: It has a disaster response mode called WiFi-Mat. It detects survivors trapped under rubble through concrete walls, classifies injury severity using START triage protocol, and estimates 3D position. The kind of tool that saves lives after earthquakes. The Rust implementation processes 54,000 frames per second. That's 810x faster than the Python version. The entire Docker image is 132 MB. The AI model fits in 55 KB of memory. Runs on an $8 ESP32 chip. Train once, deploy in any room. No retraining. No recalibration. 1,100+ tests. SHA-256 verified capability audit. 22.4K GitHub stars. 2.7K forks. MIT License. 100% Open Source.

@seunlanlege Is Polygon considered as L2 or L1? Meaning do you wait for Polygon to peg commitments onto Ethereum?

Ruthless expansion continues

LLM-Council (x.com/karpathy/statu…) vs Threshold AI Oracle (x.com/aniketpkate/st…) aka A panel of egotistic experts vs the Jury System with expert witnesses ======================= As the LLM-council by @karpathy is making rounds on X,  it is worthwhile to compare it with @SUPRA_Labs's threshold AI oracle (TAO). The key difference is that TAO wants to make the agreement/aggregation process independent of the involved LLMs/AI agents. Similar to the Jury system, TAO takes input from different LLM/AI agents, but the process of aggregation is a deterministic algorithm that is Byzantine/malicious fault-tolerant. This bring following benefits over the LLM-council. - Non-deterministic vs Deterministic: Ask the same question twice to LLM-council in a matter of minutes, the outcome will differ significantly; there is no reasoning, verifiability, or reproducibility. For TAO, for the same answers from LLM, the aggregated outcome will be exactly the same. This deterministic nature, provenance can be helpful to understand the root-cause when the system does not function as expected. For LLM-council, it can be the wild west. - Anonymity can be broken: LLM-council talks about the systems ensuring that the council members are anonymous to each other. As members see each other's responses, it is easily possible for them to make the process pseudonymous: an LLM agent may not know the identity of another agent, but it can link two or more responses by that agent to generate a pseudonymous identity. This is sufficient for agents to demonstrate adversarial behaviors towards each other: every similar to how egotistic experts may fight with each other out of pure envy or competition. - Quantifiable tolerance to Byzantine/malicious behavior: Based on the query type, TAO can determine how many wrong/missing responses it can deal with; however, as we do not understand how LLMs work well enough to quantify the fault-tolerance for the LLM-council. Looking forward to comments/critiques.

🚨 BREAKING: NYSE announces new tokenization platform. Here's what they're building: A completely new trading venue with: • 24/7 operations (no market hours) • Instant settlement (not T+1) • Stablecoin-based funding (not bank wires) • "Tokens natively issued as digital securities" Not retrofitting the existing exchange. Not adding blockchain to the back office. An entirely new venue. --- Think about what this means: NYSE will run two exchanges. The old one: 9:30-4:00 EST, T+1 settlement, bank wires. The new one: 24/7, instant settlement, stablecoin rails. They're not choosing between traditional and digital. They're operating both in parallel. --- How does this compare to others? Everyone else is building infrastructure to tokenize existing assets: • DTCC tokenizes existing custodied securities • State Street tokenizes MMFs and ETFs • Nasdaq amends rules for tokenized trading alongside traditional NYSE is building a new way to bring equities on-chain AND the venue to trade them. This puts them in competition with Figure's OPEN and Superstate. Native digital issuance. Native digital trading. --- Tokenized stocks enable a world where: • Settlement happens on-chain • Custody lives in wallets, not DTCC • Trading never stops • Capital formation happens in stablecoins The question for every institution: Are you digitizing your existing business or building the business that replaces it? NYSE just answered: both. --- #fintech #tokenization #infrastructure #digitalassets #stablecoins

@raopreetam_ @BuildOnLumia @babylonlabs_io @SUPRA_Labs @avax @tbvxyz Great to connect with you in this vibrant event of #BinanceBlockchainWeek2025

Hi, @BuildOnLumia, @babylonlabs_io, @SUPRA_Labs, @avax and @tbvxyz. 👀 📍Binance Blockchain Week. #BinanceBlockchainWeek2025.

Start of Supra Multi VM era 🎉 EVM devs welcome Move devs welcome

Thanks @haknkskn for standing with Supra. NO FUNDS FREEZING from SUPRA . Decentralisation is the core philosophy and is at the heart of blockchain thesis and revolution. We, at Supra, truly believe this and have internalised it thoroughly. In this light we understand and appreciate ByBit Lazarus Security Lab's good effort in researching 150+ blockchains and bringing out this latest report chronicling the blockchains that can actually freeze user funds. We are with ByBit team in this effort. Full report: assets.contentstack.io/v3/assets/bltf… This report mentions that Supra supports private blacklisting via config files. Shocked and surprised by this, we followed the referenced GitHub link and found that this code is NOT included in the Supra blockchain consensus or executor implementation. We have forked Aptos' aptos-core repository but are mainly using the MoveVM executor. We have our own mempool and consensus protocols. We are NOT using the code shown by this GitHub link at all. So, we affirm that there is NO account freezing logic anywhere in our system. We are communicating with Bybit Lazarus Security Lab to collaborate and requesting them to consider revising the report to omit our name with a clarification. Will keep you posted. Security and decentralisation are the foremost in the many things that are being baked in Supra cafe😋 - Encrypting transactions for MEV protection and censorship resistance - Autonomous mitigation of toxic MEV mitigation. Please listen to @JoshuaTobkin Permissionless IV talk youtu.be/gM0WbziT8Is?si… - Decentralising DORA (oracle) & dVRF nodes - We have a trustless bridge from Ethereum to Supra and we are soon going live with the other direction. We are building with the main ideals of decentralised trust.

The Bybit report revealing that 16 blockchains can freeze user funds is concerning but it also raises valid questions. Some networks built “freeze functions” for emergency recovery or exploit mitigation. Still, the line between security control and centralized override is thin. As a next-gen multi-VM Layer-1, $SUPRA’s community deserves clarity on how its architecture handles such controls — if any exist at all. A transparent statement from the team explaining how Supra ensures trustless operations, oracle integrity, and censorship resistance would be both timely and confidence-boosting. In an era where “decentralization theater” is exposed, projects that prove real autonomy will earn the market’s trust. 🔍 @SUPRA_Labs @JoshuaTobkin @JonEvansJones

Thanks @Mediitrade for pointing out. NO FUNDS FREEZING from SUPRA Decentralisation is the core philosophy and is at the heart of blockchain thesis and revolution. We, at Supra, truly believe this and have internalised it thoroughly. In this light we understand and appreciate ByBit Lazarus Security Lab's good effort in researching 150+ blockchains and bringing out this latest report chronicling the blockchains that can actually freeze user funds. We are with ByBit team in this effort. Full report: assets.contentstack.io/v3/assets/bltf… This report mentions that Supra supports private blacklisting via config files. Shocked and surprised by this, we followed the referenced GitHub link and found that this code is NOT included in the Supra blockchain consensus or executor implementation. We have forked Aptos' aptos-core repository but are mainly using the MoveVM executor. We have our own mempool and consensus protocols. We are NOT using the code shown by this GitHub link at all. So, we affirm that there is NO account freezing logic anywhere in our system. We are communicating with Bybit Lazarus Security Lab to collaborate and requesting them to consider revising the report to omit our name with a clarification. Will keep you posted.

Hey @JoshuaTobkin the Bybit report listed #Supra among chains that can freeze funds. Can you please clarify what "protocol-embedded controls" actually mean and how they stay true to decentralization?

Heyyy @hosseeb We met at permissionless but our data wasn’t in yet because Monad hadn’t released their build. At @SUPRA_Labs we already beat Monad’s consensus with our Moonshot. Our latest Hydrangea beats our original Moonshot consensus — so yeah it runs much faster than MonadBFT too. This part is key — > And now our SupraBTM outperforms their parallel execution strategy significantly as well. Posted a bounty on the matter, have your devs run it and try. We’ll increase the bounty to $40,000 usd. Here are the rules: docs.supra.com/network/evm/su… P.s. data dissemination and consensus completely dominate e2e latency compared to parallel execution — we win there already. We actually beat all web3 consensus algos, that’s why even @EspressoSys is using our Hydrangea! Ask @jillgun Moreover, yes our EVM parallel execution is also 50-100% faster as well. Here to respectfully report indeed they are not the fastest EVM L1 tech in the world — Supra has the lead here. And we did it with 20% their budget.

Watch out for the talk by @giuliano_losa on our paper, Optimistic, Signature-Free Reliable Broadcast and Its Applications, at @acm_ccs 2025. We introduce a signature-free Reliable Broadcast (RBC) protocol that achieves 2-step termination under optimistic conditions (vs. 3-step baseline) while tolerating f < n/3 Byzantine faults. Our technique extends to AVSS/AVID and also to Sailfish++, a signature-free counterpart of Sailfish and matching its latency under optimistic conditions. with @nibeshrestha2, Qianyu Yu, @giuliano_losa, @aniketpkate, @wang_xuechao

We couldn't prove it wrong. Supra wins. Block 14001207: 9.50ms → 2.44ms = 3.89× FASTER! 🔥 Block 14001274: 6.96ms → 1.85ms = 3.76× FASTER! Block 14001137: 7.54ms → 2.11ms = 3.57× FASTER! Block 14001168: 4.58ms → 1.30ms = 3.53× FASTER! Block 14001065: 14.13ms → 5.25ms = 2.69× FASTER! Average Speedup BTM over Seq. Execution: ~2.8× (PERFECT!) 💯 KEY TAKEAWAYS 1. 60% of blocks achieve 3-4× speedup 2. 90% of blocks achieve 2-4× speedup 3. ZERO blocks slower than sequential" #SupraEVM #Monad #sup

Hi CZ, we met in Taiwan a few years ago. You probably don't remember me or @JonEvansJones, but I'd like to start by saying I think what you have done for the industry is actually underappreciated. But I also think the Binance listing process needs a bit of a makeover. For example, I know you're a huge science geek and supporter. Yet, @SUPRA_Labs can't get listed even after talking to folks for years now -- we're literally one of the best, pioneering Web3 Computer Science teams in the world! The proof is that we're accepted to many of the top computer science venues globally, such as the #1 in the world, ACM CCS 2025 (we've been accepted multiple times). We have the fastest EVM parallel execution algo in the world , we have the most performant DAG design in the world (arxiv.org/abs/2505.02761) and also our Hydrangea Consensus algorithm is likewise the fastest of its type in the world (eprint.iacr.org/2025/1112). We were the 3rd Move Lang L1 to ever launch, but the Binance team listed another less technically skilled Move L2 (which now is a L1 apparently) over us, who have, to be frank, only the ability to fork technology... not make it... and it seems like their founders left the project. As you can imagine, I'm just confused about the due diligence process, how that project could get listed but we couldn't. Any pointers or suggestions? We put a majority of our resources into pushing comp sci forward and continue to do so. If the listing process is about merit, I assure you we have all the fundamentals. If it's about hype, well our DNA is more around proving our ideas than over exaggerating them for hype... but we'll work on that too. Despite how I personally feel about CEX vs. DEX, which apparently is the same how you feel, our community without any surprises wants a Binance listing. We'd love to continue to invest all we can to innovate Web3 and push the limits for our industry as a whole. Competition is good! As a lover and supporter of the sciences too, I hope this message can be seen by folks in your orbit... As a founder... I can resonate with some of the things being said regarding the listing process lately... Some of us just can't afford the ask though, especially since L1 projects like ours have block rewards that are required for long-term sustainability. The ask, fwiw, was more than founders own... To my community, the truth remains, projects like ours need more liquidity and exposure, and Binance, like it or not, is still #1 in the world... I know this too. To @cz_binance , I know you're not in charge of listings and such, but it would be cool if merit for core innovations (like being the best in the world in several protocol designs) could count for more in the decision-making process. In order for us to have innovated what we have... it requires a tremendous amount of focus, energy, and creativity (aka resources). We want to put our resources fully towards those objectives. As a founder, you know projects need to keep as much of the limited resources they have to continue to give back to Web3, discovering what is possible for decentralized networks. In short, we can't really afford the amount of tokens asked for from us either. And would love to see how we can move forward through our merits and contributions. We are one of the most sincere and serious projects in the space. What we've been able to accomplish compared to competitors 4-8X better funded than us is remarkable. Have a great day!

CCS is the best academic venue out there. Supra’s work is highlighted multiple times! Was recently told: “Supra is the smartest blockchain out there that the masses don’t know about yet.” True, this should change soon enough

@SashaSpiegelman Key insights + Original research + Scientific Validation with Peer Reviewed Conferences + Robust Engg Products + Comparison + Competition + Leading the industry is Supra. Face it 💪

This is hilarious🤡 They are comparing one implementation of BlockSTM to another 🤣🤣🤣 It took them 2 years to copy v1 - our intern could do it faster. We are releasing v2 BTW. Try to catch up.

$2.5k bounty up for grabs. If no one can prove there is a better EVM parallel execution strategy (fastest to execute while being provably correct), then the person that has the most comprehensive report, data, and tests that helps make our case will win the prize!

Attending @SurgeFintech #Dubai #Blockchain Keen on networking.